[
https://issues.apache.org/jira/browse/ARROW-1584?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16209157#comment-16209157
]
ASF GitHub Bot commented on ARROW-1584:
---------------------------------------
GitHub user buhrmann opened an issue:
https://github.com/apache/arrow/issues/1208
Serializing list arrays containing only empty lists:
(ArrowNotImplementedError: Unable to convert type: null)
Hi,
Not sure if this is related or the same as ARROW-1584, but I can't seem to
find a way to handle arrays of lists which occasionally consist of empty lists
only.
To reproduce:
```
na = [] # None, [""]
arrays = {
'c1': pa.array([["test"], na, na], type=pa.list_(pa.string())),
'c2': pa.array([na, na, na], type=pa.list_(pa.string())),
}
rb = pa.RecordBatch.from_arrays(list(arrays.values()), list(arrays.keys()))
df = rb.to_pandas()
pa.serialize_pandas(df)
# > ArrowNotImplementedError: Unable to convert type: null
tbl = pa.Table.from_pandas(df)
sink = pa.BufferOutputStream()
writer = pa.RecordBatchFileWriter(sink, tbl.schema)
writer.write_table(tbl)
# > ArrowNotImplementedError: Unable to convert type: null
```
In my use case I'm processing data in batches where individual fields
contain lists of strings. Some of the batches may, however, contain empty lists
only. And there doesn't seem to be any representation in Arrow at the moment to
deal with this situation.
Also, since I'm serializing the batches into a single file/stream, their
schemas need to be consistent, which is why I tried explicitly specifying the
type of the array as list_(string). The only workaround I've found is to
replace empty lists with [""], but that implies lots of unnecessary glue code
on the client side. Is there a better workaround until this is fixed in an
official conda release?
Many thanks,
Thomas
----
----
> [PYTHON] serialize_pandas on empty dataframe
> --------------------------------------------
>
> Key: ARROW-1584
> URL: https://issues.apache.org/jira/browse/ARROW-1584
> Project: Apache Arrow
> Issue Type: Bug
> Components: Python
> Affects Versions: 0.7.0
> Reporter: Tom Augspurger
> Assignee: Wes McKinney
> Priority: Minor
> Labels: pull-request-available, python, serialization
> Fix For: 0.8.0
>
>
> This code
> {code:python}
> import pandas as pd
> import pyarrow as pa
> pa.serialize_pandas(pd.DataFrame())
> {code}
> Raises
> {code}
> ---------------------------------------------------------------------------
> ArrowNotImplementedError Traceback (most recent call last)
> <ipython-input-71-ad21add45f0d> in <module>()
> ----> 1 pa.serialize_pandas(pd.DataFrame())
> ~/Envs/dask-dev/lib/python3.6/site-packages/pyarrow/ipc.py in
> serialize_pandas(df)
> 158 sink = pa.BufferOutputStream()
> 159 writer = pa.RecordBatchStreamWriter(sink, batch.schema)
> --> 160 writer.write_batch(batch)
> 161 writer.close()
> 162 return sink.get_result()
> pyarrow/ipc.pxi in pyarrow.lib._RecordBatchWriter.write_batch
> (/Users/travis/build/apache/arrow-dist/arrow/python/build/temp.macosx-10.6-intel-3.6/lib.cxx:59238)()
> pyarrow/error.pxi in pyarrow.lib.check_status
> (/Users/travis/build/apache/arrow-dist/arrow/python/build/temp.macosx-10.6-intel-3.6/lib.cxx:8113)()
> ArrowNotImplementedError: Unable to convert type: null
> {code}
> Presumably {{pa.deserialize_pandas}} will need a fix as well.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)